It is now well established that chronic infiltration of leukocytes into neoplastic tissue potentiates solid tumor development Many experimental and clinical studies have revealed that tumor-promoting leukocytes regulate essentially all aspects of cancer development. However, the molecular natures of individual leukocyte subtypes associated with developing breast cancers (BCs) are inadequately understood. Our overarching hypothesis is that leukocyte biomarkers in human BC can provide prognostic and predictive information regarding BC outcomes, and that these can also be utilized as platforms for design of new, individualized diagnostic and therapeutic agents. The goals of Project 1 are to further define leukocyte subtypes correlating with aggressiveness and/or specific molecular subtypes of BC and recurrent disease (Project 1, Aim 1). Identify leukocyte biomarkers that predict long-term outcome and response to therapy for patients with BC (Project 2, Aim 2). Define molecular/cellular tumor-promoting activities of high risk leukocytes in mouse models of BC to enable biomarker and target validation for anticancer therapy (Project 1, Aim 3), We will leverage human BC samples and existing murine models to define immune-stromal biomarkers in tissues that correspond to tumor-supportive or tumor-rejecting environments. Predictive/prognostic biomarkers, using our access to and expertise in complementary mouse and human BC samples, will be defined in three ways. First we will use the subdivision of macrophage and dendritic cell populations based on surface expression and/or tumor localization to define novel genetic signatures for prognosis that can then be translated into protein reagents (i.e. antibodies). We will further subdivide leukocyte infiltration using protein reagents that detect activation states for leukocytes in the microenvironment. Finally, throughout, we will track prognosis versus marker expression for a variety of BC subtypes to determine whether biomarkers sort by BC subtype.